Semantische Informationsextraktion in medizinischen Informationssystemen

Zusammenfassung Dieser Artikel beschreibt einige Erfahrungen und typische Problemstellungen mit Textmining in der Medizin und gibt einen Einblick in aktuelle und zukünftige Herausforderungen in Forschung & Entwicklung. Interessant ist n?mlich, dass auch im ,,Multimedia-Zeitalter“ die meiste Information immer noch als ,,Text“ vorliegt. Mithilfe von statistischen und linguistischen Verfahren wird mit sogenannter ,,Textmining-Software“ versucht, aus Freitexten Information ,,heraus zu schürfen“ (deshalb ,,Textmining“). Allerdings ist es damit noch nicht genug. Der n?chste Schritt besteht darin, die Information sowohl nutzbar als auch brauchbar zu machen. Die jeweiligen End-Benutzerinnen und End-Benutzer müssen in die Lage versetzt werden, auf der Basis der gewonnenen Information deren Wissen zu erweitern. In unserem konkreten Fall sollen damit Entscheidungen im Rahmen ?rztlichen Handelns unterstützt werden. Probleml?sungen in diesem Bereich erfordern eine holistische Sicht- und Herangehensweise. Daher wird es immer wichtiger, Erkenntnisse aus Informatik und Psychologie zusammenflie?en zu lassen und auf systemischer Ebene technologisch umzusetzen.     

Pharmacokinetic and bioequivalence study of two gemfibrozil preparations

A comparative pharmacokinetic study has been performed in 19 healthy male volunteers in a single-dose, randomized, two way cross-over design with two preparations of gemfibrozil (CAS 25812-30-0) capsules each of them containing 300 mg active ingredient. The test preparation was Innogem 300 mg capsule. The plasma concentration of gemfibrozil was determined by a validated HPLC-UV analytical method. The statistical comparison of individual pharmacokinetic parameters (AUC0-16, AUC0-oc Cmax, tmax) of the two capsule preparations was performed by three-way analysis of variance (ANOVA), Wilcoxon's, Westlake's, Schuirmann's and Hanck-Anderson's method as well as by the calculation of confidence intervals on the ratio of test/reference. The relative bioavailability of the test preparation with respect to the reference preparation in terms of the AUC0-oc was 104.06 +/- 21.61%. No statistically significant difference was found between the pharmacokinetic parameters, calculated from plasma concentration-time curves, indicating that the two preparations were bioequivalent.     

Optical and structural properties of MOVPE grown GaxIn1−xAs/InP strained multiple quantum well atructures

This paper presents a study of the structural and optical properties of strained GaInAs/ InP multiple quantum well (MQW) structures fabricated by LP-MOVPE. The composition of the Ga _x In_1−x As films ranged fromx = 0.17 tox = 1.0 and was determined by sputtered neutral mass spectrometry (SNMS) on thick layers. The structures of the MQW samples with well widths from 1.5 to 5 nm were investigated by high resolution x-ray diffraction (HR-XRD). Simulations of the diffraction patterns showed that transition layers of approximately 2 monolayer (ML) thickness with high lattice mismatch exist at the interfaces. Photoluminescence (PL) measurements indicate well widths of a multiple of a monolayer with local variations of one monolayer. The PL peak energies vary smoothly with the Ga concentration. These results were confirmed by optical absorption measurements.     

Microwave cavity perturbation technique: Part I: Principles

This report reviews the analysis used to extract the complex conductivity of a compound from a microwave cavity perturbation measurement. We intend to present a generalized treatment valid for any spheroidally shaped sample of arbitrary conductivity which is placed at either the electric or magnetic field antinode of the cavity. To begin with, we establish the relationship between the measured parameters and the conductivity for a spherical sample. Next, we extend these results to the case of spheroids; and for the first time, we cover all different configurations that one can possibly use to study an arbitrary conducting sample inside a cavity: in particular, all possible orientations of the sample with respect to the applied field are solved.     

Formation of droplets on Si surfaces bombarded by In from a capillary type liquid metal ion source

Droplet formation on Si surfaces bombarded by 10 keV beams from a liquid metal indium ion source has been observed. It has been verified that excess indium is transported in the ion beam, probably in the form of charged droplets. SIMS, SEM and RBS techniques have been used to study the distribution of droplets at the surface. At the specified ion source operating conditions, equilibrium droplet coverage is of the order of 25%. Individual droplets have an average diameter of 0.35 ωm and an average height of 2.2 nm. At a bombardment current density of 0.1 mA cm^?2, equilibrium coverage is reached in about 1 s.     

Computations of Explosive Boiling in Microgravity

Dynamics of the explosive growth of a vapor bubble in microgravity is investigated by direct numerical simulation. A front tracking/finite difference technique is used to solve for the velocity and the temperature field in both phases and to account for inertia, viscosity, and surface deformation. The method is validated by comparison of the numerical results with the available analytical formulations such as the evaporation of a one-dimensional liquid/vapor interface, frequency of oscillations of capillary waves, and other numerical results. Evolution of a three-dimensional vapor nucleus during explosive boiling is followed and a fine scale structure similar to experimental results is observed. Two-dimensional simulations yield a similar qualitative instability growth.     

Efficient reconstruction from non-uniform point sets

We propose a method for non-uniform reconstruction of 3D scalar data. Typically, radial basis functions, trigonometric polynomials or shift-invariant functions are used in the functional approximation of 3D data. We adopt a variational approach for the reconstruction and rendering of 3D data. The principle idea is based on data fitting via thin-plate splines. An approximation by B-splines offers more compact support for fast reconstruction. We adopt this method for large datasets by introducing a block-based reconstruction approach. This makes the method practical for large datasets. Our reconstruction will be smooth across blocks. We give reconstruction measurements as error estimations based on different parameter settings and also an insight on the computational effort. We show that the block size used in reconstruction has a negligible effect on the reconstruction error. Finally we show rendering results to emphasize the quality of this 3D reconstruction technique.     

A questionnaire study among 171 medical candidates enrolled in a PhD program. Three elements in the PhD education: supervision, research courses and international relations

One hundred and seventy-one medical doctors (median age 34 years) registered as Ph.D.-students at the Medical Faculty, University of Aarhus, were given a questionnaire concerning the Ph.D-program (91% reply rate). The Ph.D.-students had typically graduated four years before enrollment and had gained basic clinical experience. Eighty-four percent had been involved in research projects prior to their formal research education. In general, the Ph.D.-students found the supervision offered by senior researchers adequate, although, more Ph.D.-students in clinical than in preclinical departments would have liked their main supervisor to be more enthusiastic and have more specific expertise. By tradition, the Medical Faculty in Aarhus offers a broad introductory course on research methodology, this was appreciated by the Ph.D.-students. However, they found that too much time was allocated for this purpose. The Ministry of Education recommends that Ph.D.-students gain experience from international collaboration, preferably from a stay abroad. However, only 24% of Ph.D.-students had stayed at an international collaborating institution. Although the overall evaluation of the medical Ph.d.-program was positive, the Ph.D.-students pointed out weaknesses and conflicts requiring adjustment.     

The Dübendorf Study: a population-based investigation on normal values of blood pressure self-measurement

In a retrospective study of 50 consecutive children with posterior fossa tumors treated at Texas Children's Hospital, Houston, Tex., in 1989-1992, we evaluated perioperative factors which might influence the development of postoperative cerebrospinal fluid leaks. Factors analyzed included the presence of preoperative hydrocephalus, the institution of cerebrospinal fluid diversion, and the method of dural closure. No statistically significant impact on subsequent cerebrospinal fluid leakage was demonstrated.     

Boosting MUC extraction in unsatisfiable constraint networks

One very fertile domain of applied Artificial Intelligence is constraint solving technologies. Especially, constraint networks that concern problems that can be represented using discrete variables, together with constraints on allowed instantiation values for these variables. Every solution to a constraint network must satisfy every constraint. When no solution exists, the user might want to know the actual reasons leading to the absence of global solution. In this respect, extracting mucs (Minimal Unsatisfiable Cores) from an unsatisfiable constraint network is a useful process when causes of unsatisfiability must be understood so that the network can be re-engineered and relaxed to become satisfiable. Despite bad worst-case computational complexity results, various muc-finding approaches that appear tractable for many real-life instances have been proposed. Many of them are based on the successive identification of so-called transition constraints. In this respect, we show how local search can be used to possibly extract additional transition constraints at each main iteration step. In the general constraint networks setting, the approach is shown to outperform a technique based on a form of model rotation imported from the sat-related technology and that also exhibits additional transition constraints. Our extensive computational experimentations show that this enhancement also boosts the performance of state-of-the-art DC(WCORE)-like MUC extractors.